fake PC supplies

[Adam. Seychell]

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

 

[Bill Bailley]

"Adam. Seychell" <invald@invalid.com> wrote in message
news:4291c25f$0$10302$afc38c87@news.optusnet.com.au...

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at 35Amps
yet the output inductor winding is 2x1.2mm diam wire and the rectifiers
for the 5V is a 15Ax2 Schottky device. This is typical for these very
cheap power supplies. Similarly the 12V @ 18A output uses 2x8A diode
(STPR1620).

I've also seen several other PC supplies that have completely omitted EMC
components, and simply used wire links where the CM inductors are meant to
go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

They can, and they will, until these underrated and dangerous PSUs generate
sufficient fires/casualties/deaths to trigger a judicial/coroners enquiry. A
negative finding will cause ALL the companies who make the PSUs to
disappear, and the importers/resellers to wave their certified
performance/compliance paperwork as a shield against prosecution.

Money IS the root of all evil!

 

[Frithiof Andreas Jensen]

"Adam. Seychell" <invald@invalid.com> wrote in message
news:4291c25f$0$10302$afc38c87@news.optusnet.com.au...

How do they get away with this ?

Set up a Holding Company, the Holding Company Import loads of chinese
ripoffs and sell them to a group of Distributers in the form of Limited
Companies, who, in turn, will put whatever sticker on that sell the unit -
nevertheless they do not make much profit because the markup is with the
Holding Company.

Eventually, someone/something gets hurt and Distributor gets sued;
Distributor goes bankrupt; Holding Company gets to keep the profit because
*it* never did anything wrong, it just Imported the stuff.


*WHY* They can get away with this is the Interesting Question!

 

[Kelvin Chu]

Hmmm, would you have brands in mind that are bad/good?

So, should we be buying those Antec Power Supplies (TruPower or soemhting
the name is I think)?
But they are so expensive!!! (well, 2-3 times cost of cheap Power Supply...)

Kelv


"Adam. Seychell" <invald@invalid.com> wrote in message
news:4291c25f$0$10302$afc38c87@news.optusnet.com.au...

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

 

[R Adsett]

In article <4291ccf8$0$5181$afc38c87@news.optusnet.com.au>, JustMe@Home
says...

"Adam. Seychell" <invald@invalid.com> wrote in message
news:4291c25f$0$10302$afc38c87@news.optusnet.com.au...
How do they get away with this ?


They can, and they will, until these underrated and dangerous PSUs generate
sufficient fires/casualties/deaths to trigger a judicial/coroners enquiry. A
negative finding will cause ALL the companies who make the PSUs to
disappear, and the importers/resellers to wave their certified
performance/compliance paperwork as a shield against prosecution.

Money IS the root of all evil!

The love of money is the root of all evil!

The quote makes more sense when it's complete. Also it's "The proof of
the pudding is in the tasting" not "The proof is in the pudding"

Robert (feeling pedantic today)

 

[Rich Grise]

On Mon, 23 May 2005 21:45:35 +1000, Adam. Seychell wrote:

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

By using insanely high switching frequencies. Ten or fifteen years ago,
I had a client who wanted to build a 24V, 30A switcher for an on-board
wheelchair battery charger; three months later we abandoned the project,
because even the guys from the company who sold the magical chip weren't
able to teach me switcher design. The chip was rated for 1 MHz. A
MEGAHERTZ! In a SWITCHER! =:-O These days, I wouldn't be surprised if
they're approaching gigahertz switching frequencies - heck, look at
microprocessors these days, that are big enough to be their own antenna!

Have you seen any of these magical mysterious power supplies kill anybody,
or even let the magic smoke out?

Thanks,
Rich

 

[Pooh Bear]

"Adam. Seychell" wrote:

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

a) Probably because no-one actually *needs* a 500W power supply and the
dodgy suppliers know this.

b) Because it's possible to cheat and lie. Enron did this rather
dramatically for example.

Graham

 

[Rich The Philosophizer]

On Mon, 23 May 2005 22:30:44 +1000, Bill Bailley wrote:

Money IS the root of all evil!

WRONG!!! WRONG!!! WRONG!!! WRONG!!! WRONG!!!!!!!!!!!!!!!!

Money is merely a scorekeeping device. It's inanimate. It just sits
there, until somebody uses it to buy something. My Mom, bless her, was
very specific in teaching me that it's the LOVE of money that's the root
of all evil.

LOVE of money is the root of all evil.

That's LOVE of money as a replacement for love of life, love of
pepole, love of love, love of reality, and so on.
--
Love,
Rich Grise
First Member, Church of the Neodruid

In truth, it's denial that's the root of all evil, but that's advanced
metaphysics - for further information, please visit
http://www.godchannel.com

 

[Spehro Pefhany]

On Mon, 23 May 2005 16:47:37 +0100, the renowned Pooh Bear
<rabbitsfriendsandrelations@hotmail.com> wrote:

"Adam. Seychell" wrote:

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

a) Probably because no-one actually *needs* a 500W power supply and the
dodgy suppliers know this.

b) Because it's possible to cheat and lie. Enron did this rather
dramatically for example.

Graham

Tom's Hardware has tested some PC power supplies.

http://www6.tomshardware.com/howto/20021021/powersupplies-07.html

---
Inadequate and Deceptive Product Labeling:
Comparison of 21 Power Supplies

Test Results In Detail

Let's not beat about the bush here. The 21 power supplies we tested
are rated by their manufacturers at between 300 and 520 watts. Our
tests showed, however, that only 15 of them actually met their
specifications....

---

Of course the good ones can cost more than a whole cheap PC...



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Guy Macon]

Spehro Pefhany wrote:

Tom's Hardware has tested some PC power supplies.

http://www6.tomshardware.com/howto/20021021/powersupplies-07.html

---
Inadequate and Deceptive Product Labeling:
Comparison of 21 Power Supplies

Test Results In Detail

Let's not beat about the bush here. The 21 power supplies we tested
are rated by their manufacturers at between 300 and 520 watts. Our
tests showed, however, that only 15 of them actually met their
specifications....

Many years ago I had a no-name PC Power supply take out a motherboard,
both hard disks, all the cards, and even the keyboard. Since then
I have never powered up a new system without first replacing the PS
with one from PC Power and Cooling, which I consider to be the maker
of the best PC power supplies available.

Of course the good ones can cost more than a whole cheap PC...

...and my data is worth more than both put together.

 

[Anthony Fremont]

"R Adsett" wrote:

JustMe@Home wrote:

Money IS the root of all evil!

The love of money is the root of all evil!

The quote makes more sense when it's complete. Also it's "The proof
of
the pudding is in the tasting" not "The proof is in the pudding"

My favorite misquote is "Pride goeth before a fall". The correct phrase
is "Pride goeth before destruction, and an haughty spirit before a fall"
(Proverbs 16:18).

 

[Spehro Pefhany]

On Mon, 23 May 2005 17:32:20 +0000, the renowned Guy Macon
<_see.web.page_@_www.guymacon.com_> wrote:

Spehro Pefhany wrote:

Tom's Hardware has tested some PC power supplies.

http://www6.tomshardware.com/howto/20021021/powersupplies-07.html

---
Inadequate and Deceptive Product Labeling:
Comparison of 21 Power Supplies

Test Results In Detail

Let's not beat about the bush here. The 21 power supplies we tested
are rated by their manufacturers at between 300 and 520 watts. Our
tests showed, however, that only 15 of them actually met their
specifications....

Many years ago I had a no-name PC Power supply take out a motherboard,
both hard disks, all the cards, and even the keyboard. Since then
I have never powered up a new system without first replacing the PS
with one from PC Power and Cooling, which I consider to be the maker
of the best PC power supplies available.

Even if you don't get the fireworks you can get crashes, data
corruption etc.. See the above article which had one supply dropping
the 12V line to < 6V. I'm sure the typical $99 200GB drive is going to
love that.

Of course the good ones can cost more than a whole cheap PC...

..and my data is worth more than both put together.

Yes, even the hassles of restoring a system that has fried itself is
probably worth more. Assuming you have full backups.


Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Mike Monett]

Guy Macon wrote:

[...]

Many years ago I had a no-name PC Power supply take out a motherboard,
both hard disks, all the cards, and even the keyboard. Since then
I have never powered up a new system without first replacing the PS
with one from PC Power and Cooling, which I consider to be the maker
of the best PC power supplies available.

Of course the good ones can cost more than a whole cheap PC...

..and my data is worth more than both put together.
I had the same thing happen. Fortunately I keep good backups so I didn't

lose anything.

But even the best power supply can fail, or an external device could
develop a short between voltages. I've been thinking of a simple crowbar
that could be added to the +12V and +5V supplies. Have it short the
voltage if a failure raises the voltage above a limit, and plug it in a
spare drive power connector to prevent damage if anything goes wrong.

I don't know how power supplies react to a short. If they respond by
trying to increase the other voltages, it mmight be necessary to crowbar
all the voltages at the same time. I'd probably do it just to be on the
safe side.

One problem would be indicating which voltage triggered the circuit so
you could troubleshoot it. This might take a small battery powered
circuit to store the conditions at failure, and a switch to light the
corresponding led to examine the failure.

This could be made real cheap. Anyone know a source for inexpensive 50A
or 100A scr's?

Mike Monett

 

[Spehro Pefhany]

On Mon, 23 May 2005 15:28:42 -0400, the renowned Mike Monett
<no@spam.com> wrote:

Guy Macon wrote:

[...]

Many years ago I had a no-name PC Power supply take out a motherboard,
both hard disks, all the cards, and even the keyboard. Since then
I have never powered up a new system without first replacing the PS
with one from PC Power and Cooling, which I consider to be the maker
of the best PC power supplies available.

Of course the good ones can cost more than a whole cheap PC...

..and my data is worth more than both put together.
I had the same thing happen. Fortunately I keep good backups so I didn't
lose anything.

But even the best power supply can fail, or an external device could
develop a short between voltages. I've been thinking of a simple crowbar
that could be added to the +12V and +5V supplies. Have it short the
voltage if a failure raises the voltage above a limit, and plug it in a
spare drive power connector to prevent damage if anything goes wrong.

Some (most?) PC power supplies already have crowbar or some other
kind of overvoltage protection built in. For example (this is a beefy
server supply):

http://www.enermax.com.tw/download/EPS_12V_English.pdf

See table 3.5.2.

Here's another:

http://www.antec.com/specs/TPII430_spe.html

In fact it's required in the Intel's Design Guide:

http://www.formfactors.org/developer%5Cspecs%5CATX_ATX12V_PS_1_1.pdf

See section 3.4, although the Enermax supply does not meet the
overvoltage specification on the 3.3V line.

I always liked section 3.1.5 as well. What exactly constitutes
"excessive smoke", I wonder. ;-)

I don't know how power supplies react to a short. If they respond by
trying to increase the other voltages, it mmight be necessary to crowbar
all the voltages at the same time. I'd probably do it just to be on the
safe side.

One problem would be indicating which voltage triggered the circuit so
you could troubleshoot it. This might take a small battery powered
circuit to store the conditions at failure, and a switch to light the
corresponding led to examine the failure.

This could be made real cheap. Anyone know a source for inexpensive 50A
or 100A scr's?

Teccor has some 50A SCRs in TO-3P for about $5 US each one-off (only
about $1.70 for 40A in TO-220, so much better value). Much bigger and
you're probably into stud package and more money.



Best regards,
Spehro Pefhany
--
"it's the network..." "The Journey is the reward"
speff@interlog.com Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog Info for designers: http://www.speff.com

 

[Franc Zabkar]

On Mon, 23 May 2005 21:45:35 +1000, "Adam. Seychell"
<invald@invalid.com> put finger to keyboard and composed:

I'm recently seeing more cheap PC supplies on the market which appear to
be fake, in that the clammed output power not impossible for the size of
parts inside. e.g I have a 500W "Shaw" brand ATX supply, and its output
inductor uses a piddly T106 (26mm OD) core. The 5V is specified at
35Amps yet the output inductor winding is 2x1.2mm diam wire and the
rectifiers for the 5V is a 15Ax2 Schottky device. This is typical for
these very cheap power supplies. Similarly the 12V @ 18A output uses
2x8A diode (STPR1620).

I've also seen several other PC supplies that have completely omitted
EMC components, and simply used wire links where the CM inductors are
meant to go. Some PC supplies I've come across have even used standard
ceramic/polyester capacitors in place of the Y and X2 rated safety
capacitors.

How do they get away with this ?

Read about my experience with a Deer 170W PSU masquerading as 400W.

http://groups-beta.google.com/group/aus.electronics/msg/1515813ff03788f0?hl=en
http://groups-beta.google.com/group/alt.comp.hardware.homebuilt/msg/e705e80f9c9611d1?hl=en


- Franc Zabkar
--
Please remove one 's' from my address when replying by email.

 

[Mike Monett]

Ian Stirling wrote:

[...]

I wonder about a 'gentler' way.
If you simply disconnect the main energy storage cap, then the supply
won't stay pumping power into the output caps for more than a few
microseconds.

Yes, I as thinking along the same line. What about a separate unit that
monitors the voltages, crowbars them in case of a fault, and removes
the ac power to the computer?

As Speff points out, modern supplies are supposed to have the crowbars
built in. However, Win 3.1 and Win 95 run fine on older computers, so
they still have plenty of useful life. For example, I use Win 3.1 for
internet access and have zero problems with trojans, viruses, popups,
etc. They won't run on my system. Also have zero problem with upgrades
breaking things. They don't exist

Mike Monett

 

[Mike Monett]

Spehro Pefhany wrote:

Some (most?) PC power supplies already have crowbar or some other
kind of overvoltage protection built in. For example (this is a
beefy server supply):

http://www.enermax.com.tw/download/EPS_12V_English.pdf

See table 3.5.2.

Here's another:

http://www.antec.com/specs/TPII430_spe.html

In fact it's required in the Intel's Design Guide:

http://www.formfactors.org/developer%5Cspecs%5CATX_ATX12V_PS_1_1.pdf

See section 3.4, although the Enermax supply does not meet the
overvoltage specification on the 3.3V line.

I always liked section 3.1.5 as well. What exactly constitutes
"excessive smoke", I wonder. ;-)

[...]

Teccor has some 50A SCRs in TO-3P for about $5 US each one-off
(only about $1.70 for 40A in TO-220, so much better value). Much
bigger and you're probably into stud package and more money.

Best regards,
Spehro Pefhany

Thanks, Speff. The Teccor 40A scr looks very good.

Although the newer supplies may already have protection, many older
computers still have plenty of use. For example, I log on the web
with a 200MHz machine running Win 3.1, and have zero problems with
viruses and trojans. They simply won't run. I use Win 95 on a 166MHz
machine to view pdf files that need Acrobat 4, and my main computer
is 450MHz running Linux. Each computer needs a backup, so I am
looking at a triple boot machine running 166MHz. I also use several
other cpu's for process control.

This gives a number of machines without protection, and I have
already lost two due to power supply failure. So I'm thinking of a
small ac-powered unit that crowbars the voltages and removes the AC
to a faulty supply. A 40A scr probably won't even need a heat sink

The Intel failure section is hilarious. I copied it below for those
who don't have the time to download it:

--------------------------------------------------------------------
3.1.5 Catastrophic Failure Protection

Should a component failure occur, the power supply should not
exhibit any of the following:

- Flame
- Excessive Smoke
- Charred PCB
- Fused PCB conductor
- Startling noise
- Emission of molten material
--------------------------------------------------------------------

They forgot "Emits funny odor".

Mike Monett

 

[Pooh Bear]

Mike Monett wrote:

The Intel failure section is hilarious. I copied it below for those
who don't have the time to download it:

--------------------------------------------------------------------
3.1.5 Catastrophic Failure Protection

Should a component failure occur, the power supply should not
exhibit any of the following:

- Flame
- Excessive Smoke
- Charred PCB
- Fused PCB conductor
- Startling noise
- Emission of molten material

With the exception of *startling noise* - all that is covered by IEC60950.
Or UL1950 in your case.

--------------------------------------------------------------------

They forgot "Emits funny odor".

Nah - that's 'user notification of failure'. ;-)

Graham

 

[Mike Monett]

Pooh Bear wrote:

Mike Monett wrote:

The Intel failure section is hilarious. I copied it below for those
who don't have the time to download it:

--------------------------------------------------------------------
3.1.5 Catastrophic Failure Protection

Should a component failure occur, the power supply should not
exhibit any of the following:

- Flame
- Excessive Smoke
- Charred PCB
- Fused PCB conductor
- Startling noise
- Emission of molten material

With the exception of *startling noise* - all that is covered by IEC60950.
Or UL1950 in your case.

[...]

Graham

Do you have a copy of UL1950 or IEC60950? According to this site, IEC60950
is $337.48 USD:

http://www.techstreet.com/cgi-bin/detail?product_id=930919

If you don't have copies, do you have another reference that discusses
failure protection?

Mike Monett

 

[Clifford Heath]

R Adsett wrote:

The love of money is the root of all evil!
The quote makes more sense when it's complete.

Quite. The quote is:
"The love of money is the root of all *kinds of* evil".
Quite different again... and spectacularly different
from the oft misquoted version.